extract-urls.py

#!/usr/bin/env python3
import fileinput
import re

##
# Extract URLs from file or `stdin`
#
# Prints out a standard list of url address substrings contained in `fileinput`
# Call on a file or pipe to `stdin`
#
    # extract-url.py messy_file.txt
    # ./some-script --messy | extract-url.py | sort | unique > urls.txt
##

regex = re.compile(r"https?://(?:[-\w.]|(?:%[\da-fA-F]{2}))+")

string = ""
for line in fileinput.input():
        string += line

matches = re.findall(regex, string) # => array
print("\n".join(matches))